Safety & resistance
June 2026Do lysins drive resistance like antibiotics?
“If antibiotics drive resistance, won’t your lysins do the same?”
It is the first question almost everyone asks, and it is the right one to ask. The short answer is no, and the reason is not wishful thinking. It is structural biology. Antibiotics and lysins attack bacteria in fundamentally different ways, and that difference is exactly why one reliably breeds resistance and the other does not.
In one sentence
Antibiotics block a changeable process inside the cell and take hours to act, giving bacteria time to adapt; lysins cut the unchangeable structural wall from the outside in seconds, giving them neither a target they can change nor the time to change it.
Why antibiotics breed resistance
An antibiotic has to get inside the bacterium and switch off a specific protein or process. That opens six well-mapped escape routes: bacteria can pump the drug back out, lock the doors it enters through, produce enzymes that destroy it, mutate the target so the drug no longer fits, build a drug-proof substitute, or simply swap the resistance gene with a neighbour.
Each of these changes something the cell can afford to lose. And because antibiotics kill slowly, over many bacterial generations, a single rare resistant cell has time to multiply while the drug clears the rest. Resistance to a new antibiotic is usually detectable within a year or two of widespread use.
Why lysins are different
Lysins are enzymes that bacteriophages evolved to cut open the bacterial cell wall, a mesh called peptidoglycan. That wall is the bacterium’s skeleton: a single, continuous, woven molecule holding the cell together against enormous internal pressure. A lysin snips one of its load-bearing bonds and the cell bursts in seconds, like a balloon.
Crucially, a lysin works from outside the cell and never goes in. So the six antibiotic escape routes, which all operate inside the cell or at its membrane, are pointed the wrong way. There is nothing to pump out, no door to lock, no internal target to mutate. The only theoretical escape is to rebuild the wall itself, and a bacterium cannot do that without dying.
Essential
The target can’t change
Peptidoglycan is the bacterial skeleton, conserved for around three billion years, because altering it is fatal. It cannot be swapped for a resistant version.
External
The attack is from outside
Every antibiotic defence faces inward. A lysin never enters the cell, so none of those shields apply, including against dormant ‘persister’ cells.
Fast
There’s no time to adapt
A lysin delivers a 1,000-fold kill in minutes, faster than a bacterium can divide. The window resistance needs to select a mutant never opens.
What the evidence shows
This is not only theory. In a 28-day head-to-head laboratory study, the gold standard for measuring how fast resistance emerges, a lysin and three antibiotics were pushed under identical conditions. The antibiotics became 8, 16 and 32 times harder to use. The lysin needed exactly the same dose on day 28 as on day 1: zero drift.
Fold-increase in the dose needed to stop growth, over 28 days (lower is better)
0×
Lysin (exebacase)
8×
Vancomycin
16×
Daptomycin
32×
Oxacillin
The same pattern holds across many lysins and 40+ passage cycles, in a decade of real-world surveillance, and across six lysins tested in human clinical trials, where not a single case of lysin resistance has ever been recorded. Remarkably, adding a trace of lysin alongside an antibiotic can even suppress the resistance the antibiotic would otherwise develop.
The honest position
We are careful not to overclaim, because candour is more persuasive than absolutes. There is a non-zero chance of lysin resistance. What the structural biology and the data show is that it is orders of magnitude less probable than antibiotic resistance, not that it is impossible.
Even the one peptidoglycan-targeting enzyme where resistance has been forced in the lab, lysostaphin, produces mutants that grow 4–5× slower, are at least 5× less virulent, and revert to being treatable by ordinary antibiotics. Even the worst case is a self-limiting dead end.
| We do NOT claim | We DO claim |
|---|---|
| Resistance is impossible | Resistance is extremely unlikely |
| — | Resistance has not been observed in studies |
| — | Resistance has never been clinically confirmed |
Why this matters even more for personal care
ArrowBiome’s lysins are INCI cosmetic actives applied to skin, not systemic drugs. Every factor that drives resistance in a hospital is weaker or absent in a rinse-off, species-selective topical: no lingering low dose, no exposure to the gut’s reservoir of resistance genes, and no flattening of the beneficial skin community. The use-case lowers an already low risk.
The analogy to remember
An antibiotic picks a lock the bacterium can re-key. A lysin pops the balloon. You cannot evolve resistance to losing your own skeleton, and you are not given the time to try.
Selected references
- Oh et al. 2023, Microbiology Spectrum — exebacase serial passage
- Sauvé et al. 2024, J Infect Dis — engineered Gram-negative lysin
- Fischetti 2018, Viruses — lysin development & resistance
- Kusuma et al. 2006, Antimicrob Agents Chemother — lysostaphin resistance fitness cost
- Murray et al. 2022, The Lancet — global burden of AMR
- Qian & Zheng 2026, Front Microbiol — endolysin resistance frameworks
Full citations with DOIs are included in the downloadable briefing. We grade each claim as strong, moderate or preliminary, and never claim resistance is impossible, only that it is extremely unlikely, unobserved and clinically unconfirmed.
